Retroviruses exist as infectious viruses and as endogenous retroviral copies (ERVs) which are viral DNA copies integrated into host DNA. Such ERVs are a permanent part of the host genome and represent 8-10% of host chromosomal DNA. Retroviruses were first recognized as naturally occurring infectious agents linked to various neoplasms in their host species. We have been engaged in an ongoing effort to characterize pathogenic viruses, and identify the various host factors that restrict the replication cycle of these viruses. Mouse leukemia viruses (MLVs) are gammaretroviruses linked to induction of neoplasms and to neurological and immunodeficiency diseases. Inbred strains of laboratory mice and wild mouse species differ in their susceptibility to mouse gammaretrovirus infection and to virus-induced diseases, and they also differ in the types of MLVs that they carry. Susceptibility differences are due to variations in specific host genes that restrict virus replication, and we have been engaged in an ongoing effort to identify and characterize host genes that are either involved in virus resistance or that contribute to the disease process. There are two types of host genes involved in virus-induced disease. First, the mouse genome contains copies of mouse gammaretrovirus genomes, many of which can produce infectious and pathogenic viruses. Second, there are also host factors that interfere directly with virus infection and replication, and we are particularly interested in those factors that inhibit virus entry and the early post-entry stages of the virus replicative cycle. At the level of entry, resistance can be caused by polymorphisms in the cell surface receptors. After the gammaretrovirus enters the receptive cell, reverse transcription and translocation to the nucleus can be inhibited or altered by virus resistance factors, especially Fv1, mApobec3, and TRIM5alpha. Our group aims to characterize the endogenous retroviruses carried in the mouse genome, and the host encoded resistance factors and their viral targets. Our ultimate goal is to describe co-evolutionary patterns of virus-host interactions in natural populations. This work relies heavily on wild mice because laboratory strains provide only a limited sampling of the genetic diversity in Mus. Also, wild mouse species allow us to examine survival strategies in natural populations that harbor virus and to follow the evolution of the resistance genes. These mice additionally provide a source of novel resistance genes and virus variants. One set of projects aims to characterize viruses found in pre-leukemic and leukemic tissues during spontaneous lymphomagenesis. These viruses can show three host range subtypes: ecotropic, xenotropic and polytropic mouse leukemia viruses (E-, X-, P-MLVs) which differ in their ability to infect cells of mouse and other species. Leukemogenesis involves generation of recombinants with polytropic host range and these viruses represent recombinants of ERVs derived from all three host range groups. Although P-MLVs are deemed to be the proximal agents of disease induction, few biologically characterized infectious P-MLVs have been sequenced for comparative analysis. We analyzed the complete genomes of 16 naturally occurring infectious P-MLVs, 12 of which were typed for pathogenic potential. We sought to identify ERV progenitors, recombinational hotspots, and segments that are always replaced, never replaced, or linked to pathogenesis or host range. Each P-MLV has an E-MLV backbone with P- or X-ERV replacements that together cover 100% of the recombinant genomes, with different substitution patterns for X- and P-ERVs. Two segments are always replaced, in envelope (Env): the N-terminus of the surface subunit, and the cytoplasmic tail R peptide. Viral gag gene replacements are influenced by host restriction genes Fv1 and Apobec3. Pathogenic potential maps to the env transmembrane subunit segment encoding the N-heptad repeat (HR1). Molecular dynamics simulations identified three novel interdomain salt bridges in the lymphomagenic virus HR1 that could affect structural stability, entry or sensitivity to host immune responses. The long terminal repeats of lymphomagenic P-MLVs are differentially altered by recombinations, duplications or mutations. We also analyzed X-MLVs found in disease tissue and determined that one is also a recombinant that alters the viral envelope and its entry properties. This analysis of the naturally occurring, sometimes pathogenic P-MLV recombinants defines the limits and extent of intersubgroup recombination, and identifies specific sequence changes linked to pathogenesis and host interactions. Our interest in host factors that restrict retroviruses has focused largely on the Fv1 gene, the first antivirat host gene to be discovered. The laboratory mouse Fv1 gene was first described for its ability to mediate resistance to murine leukemia viruses (MLVs). Sequence similarity between Fv1 and the gag protein of the murine endogenous retrovirusL (MuERV-L) family of ERVs suggests that Fv1 was co-opted from an ancient provirus. Previous evolutionary studies found Fv1 orthologs only in the Mus genus. We have now used genome database searches and DNA sequencing to identify orthologous Fv1 sequences in several species belonging to multiple families of rodents outside of the genus Mus. We show that these Fv1 orthologs are in the same chromosomal region in all species, between the genes Miip and Mfn2. The species distribution suggests a minimum insertion time of 45 million years for the ancient progenitor of Fv1. Our analysis also reveals that Fv1 was not detectable or heavily mutated in some lineages in the superfamily Muroidea. In concert with our previous findings in the Mus genus, we found strong evidence of positive selection of Fv1 in the African clade in the subfamily Muridae. Residues identified as evolving under positive selection include those that have been previously found to be important for restriction of multiple retroviral lineages. Taken together these findings suggest that the evolutionary origin of Fv1 substantially predates Mus evolution, that the rodent Fv1 has been shaped by lineage-specific differential selection pressures, and that Fv1 has long been evolving under positive selection in the rodent family Muridae, supporting a defensive role that significantly antedates exposure to MLVs. Another ongoing study is in the process of characterizing a novel 8.0 kb endogenous retrovirus, XTERV-LS, that we identified in the amphibian, Xenopus tropicalis. This ERV has intact open reading frames for all viral proteins, but has an unusual genomic structure and domain relationships to known retroviruses. Phylogenetic analysis failed to identify close relationships with known retroviruses and XTERV-LS is distinct from the 2 previously described X. tropicalis ERVs: XTERV-LS1 and Xen-1. The reverse transcriptase domain of this ERV shows closest similarity to the ancient envelope (env)-deficient ERV-L family of endogenous retroviruses, and to the exogenous spumaviruses confirming an evolutionary relationship between these subgroups. The C-terminal end and the TM domain of the XTERV-LS Env clusters with the mammalian syncytin genes that function in trophoblast fusion in placenta formation. This newly acquired retrovirus provides a link between the ancient ERV-L and spumaviruses as well as a connection to the retrovirus Env genes captured as syncytins. Finally, we continue to analyze a novel host restriction factor found in most mammals that blocks viral protein production in the later stage of the viral life cycle. Experimental evidence indicates that inspliced message may be the target of this restriction.